Cyclospora on the march in the US
Jumping on the 'explosive diarrhea' bandwagon
Cyclospora has been marching across the US, striking fear in all produce lovers with dramatic headlines about ‘explosive diarrhea.’ When I was working as an enteric disease epidemiologist Cyclospora was one of my least favorite pathogens to investigate, for a number of reasons.
Limited sequencing. CDC has been working hard in recent years to develop and evaluate methods, but there is not routine sequencing of samples in the same way there are with the bacterial pathogens. This means that for real-time detection of outbreaks, it’s really about the old-fashioned shoe leather epi work.
Long incubation. Though it’s not as long as some pathogens like Hepatitis A or Listeria, it can take several week, so patient interviews ask them to remember what they ate in the two weeks prior to their illness onset, which means they’re trying to remember what they ate a month ago.
Challenging diagnostics. This part has improved dramatically in the past several years. Prior to the advent of molecular diagnostic panels which test for many pathogens at once, diagnosis relied on various forms of microscopy. This meant that a provider had to suspect a parasitic infection, order the right kind of test, and even then low or intermittent shedding made it easier to miss.
Syndromic Trends
Speaking more about the diagnostics, this is a great opportunity to bring in one of my favorite syndromic surveillance sources - BIOFIRE Diagnostics syndromic trends, which can be found at https://syndromictrends.com/. Anonymized Data is aggregated from instruments in clinical labs across the country and then presented as trends of percent positivity over time. Looking at their Cyclospora data the increase is pretty dramatic, rising from 1.9% of tests positive for Cyclospora June 14-20, to 10.9% positive July 5-10th. WOW. They have data available for last summer as well, and you can see that things peaked in July at just 1.2%.
Shoe-leather epidemiology
So, in the absence of real-time comprehensive sequencing to quickly tell you which cases are likely part of an outbreak, what steps are taken to try and identify the source? The first and primary source is through patient interviews. After a lab or healthcare facility identifies a patient has tested positive for Cyclospora, they are required in most states to report that to the health department. This leads to the most critical step in solving an outbreak - patient interviews. There is a unified questionnaire for Cyclospora that asks about travel, events, grocery stores, and dozens of individual food items. This form is called the Cyclospora National Hypothesis Generating Questionnaire, or CNHGQ. These interviews are detailed, and all of the individuals who take the time to complete these are heroes in my book - outbreaks would never be solved without their willing participation.
Want to try it out for yourself? Download the full questionnaire and see how well you can fill it out for yourself - to simulate what is asked of patients, look at todays date and go back two weeks, and then fill it out based on the 14 days prior to that date. Whenever I used to teach interview training at the health department, I would have learners interview each other using the forms they would be using with the public - it helps you better understand what you’re asking of people, and why it might be difficult!
Interestingly, the CNHGQ is currently up for approval to extend approval for the use of the CNHGQ - it’s open for public comment right now!
Tracking the data
Because of the recent uptick in cases, I’ve started an ongoing tracker at genomicepi.com/outbreaks/cyclospora/. This takes the same data that is reported to the CDC and presents it in a few different ways. One of my favorites is the year-over-year chart. This can help give a sense of whether the trajectory we are seeing is within the usual range, and as you can see cases are increasing rapidly, and sooner than we’ve seen in 2022-2025.
For investigating geographically, the hexmap gives an overview where you can see at a glance that the Michigan/Ohio area is leading the way. To note, Michigan is actually reporting even more cases on their state website - this dashboard is built on what has been reported to the CDC, which typically lags behind. If you explore the live map you can click on any state to get a popup view which shows the trajectory of this year compared to previous years for that state. If you want to see some impressive slopes, go click on Michigan or Ohio! Conversely, clicking on other states like Florida shows that some are still within their historical range.
Wrap up
I’ve set the dashboard to update automatically every week when the CDC dataset gets updated, which looks to be on Thursdays. I’m expecting another big spike in confirmed cases next Thursday after that update. CDC recently updated their Cyclospora page to acknowledge the increase, and they’ll be providing more frequent updates.
One important note with the genomicepi.com dashboard - CDC focuses their dashboard on cases that have been confirmed to NOT have international travel - in other words cases where individuals both live in the US and were exposed in the US. This is an important distinction because when looking for outbreaks, we need to focus on locally-acquired cases. This data is available to CDC through the CNHGQ I mentioned earlier, but this data is not available in the public case surveillance dataset. For that reason, my dashboard shows all confirmed cases regardless of whether they were acquired domestically or internationally.
Hope this has been helpful - if there are other visuals or things you’d like to see on the dashboard let me know!